Skip to content

The Southwest and Northeast Monsoon: Onset, Withdrawal and Breaks

CSIR NET Part C meteorology: SW monsoon onset over Kerala, advance, active and break spells, withdrawal criteria, and the NE monsoon over south India.

  • 17 min read
  • Updated
On this page · 16 sections
  1. Prerequisites
  2. The Hook: Declaring the Monsoon
  3. The Monsoon Calendar
  4. The Southwest Monsoon System
  5. Onset
  6. Active Spells, Weak Spells and Breaks
  7. Withdrawal
  8. The Northeast Monsoon
  9. Interannual Variability (in brief)
  10. Memory Tricks
  11. Don’t Confuse
  12. Common Mistakes
  13. Practice
  14. Quick Revision Table
  15. Key Takeaways
  16. What’s Next

Prerequisites

The Hook: Declaring the Monsoon

How CSIR NET asks it. Part C gives you station rainfall, wind and OLR data and asks whether onset can be declared; gives a rainfall-anomaly series and asks you to count break spells; asks where it rains during a break, what changes in the jets at onset and withdrawal, and how the northeast monsoon differs from the southwest one (season, wind, regions, ENSO link).


The Monsoon Calendar

graph LR
    A["Mid-May
monsoon over the
Andaman Sea"] --> B["~1 June
onset over Kerala"] B --> C["June
advance in two branches;
mid-June hiatus common"] C --> D["~8 July
covers the
whole country"] D --> E["July-August
active and break spells"] E --> F["~17 September
withdrawal begins
from NW India"] F --> G["~15 October
withdrawal complete;
NE monsoon sets in
over Tamil Nadu"] G --> H["October-December
northeast monsoon"]
EventIMD normal date
Onset over Kerala1 June
Monsoon covers the entire country8 July
Withdrawal begins from north-west India17 September
Withdrawal from the whole countryabout 15 October

These are the normals revised by IMD in 2020; before that the whole country was covered by 15 July and withdrawal began on 1 September. The onset date over Kerala varies from year to year by about a week (one standard deviation).

Seasonal factValue
SeasonJune–September
All-India long period average (LPA)868.6 mm (about 87 cm; 1971–2020 base)
Share of annual all-India rainfallabout three-quarters
Realised season: normalwithin ±10% of LPA
Deficient / excessbelow 90% / above 110% of LPA
IMD forecast categoriesdeficient (< 90%), below normal (90–95%), normal (96–104%), above normal (105–110%), excess (> 110%)

The Southwest Monsoon System

ComponentLevelRole
Mascarene Highsurface, about 30° S, south Indian Oceansource of cross-equatorial flow
Somali (Findlater) jetabout 850 hPaconcentrated cross-equatorial current along East Africa; carries moisture into the Arabian Sea
Heat lowsurface, NW India–Pakistanwestern end of the monsoon trough
Monsoon troughsurface to mid-troposphereaxis of convergence from the heat low to the head of the Bay
Tibetan anticycloneabout 200–100 hPaupper-level outflow over the heated plateau
Tropical easterly jetabout 100–150 hPaupper easterlies over the peninsula
Low-level westerliesup to about 600 hPa over the Arabian Seathe “monsoon current” itself

Two branches. The Arabian Sea branch strikes the Western Ghats nearly at right angles (heavy orographic rain on the windward coast, rain shadow inland). The Bay of Bengal branch reaches north-east India and Myanmar, is deflected by the Arakan hills and the Himalaya, and flows west along the monsoon trough, so rainfall decreases from east to west across the Gangetic plain.


Onset

What happens physically

Before onset the Somali jet strengthens, the westerlies over the south-east Arabian Sea deepen, and organised deep convection (very low outgoing longwave radiation) moves north from the equatorial Indian Ocean. At onset the system often changes abruptly:

  • a sudden increase in kinetic energy of the low-level westerlies over the Arabian Sea;
  • a northward jump of the cloud band and the low-level jet axis;
  • sometimes an onset vortex over the south-east or east-central Arabian Sea, which can intensify into a depression or cyclone and move north or north-west, drawing the monsoon current after it;
  • in the upper troposphere, the subtropical westerly jet moves north of the Tibetan Plateau and the TEJ establishes itself, reflecting the reversal of the upper-tropospheric meridional temperature gradient over South Asia.

IMD’s criteria for onset over Kerala

IMD’s operational criteria (in use since about 2006) combine rain, wind and cloud:

ElementCriterion
Rainfallafter 10 May, if 60% of a list of 14 stations in Kerala, Lakshadweep and coastal Karnataka (Minicoy, Amini, Thiruvananthapuram, Punalur, Kollam, Alappuzha, Kottayam, Kochi, Thrissur, Kozhikode, Thalassery, Kannur, Kudulu, Mangaluru) report 2.5 mm or more for two consecutive days, onset is declared on the second day, provided the following also hold
Wind fieldwesterlies extend up to 600 hPa over the box equator–10° N, 55° E–80° E; zonal wind of about 15–20 kt at 925 hPa over 5–10° N, 70–80° E
Outgoing longwave radiation (OLR)satellite-derived OLR below 200 W m⁻² over 5–10° N, 70–75° E

Sixty per cent of 14 stations is 8.4, so in practice at least 9 stations must report.

Advance

After onset over Kerala the monsoon advances northward in surges:

  • the Arabian Sea branch along the west coast;
  • the Bay of Bengal branch into north-east India and then westward along the Gangetic plain;
  • the two meet over north-west India, and the monsoon covers the whole country.

A hiatus (stall) in the advance, often in mid-June, is common. Each surge is usually tied to a northward-propagating convective band or a low or depression forming at the head of the Bay.


Active Spells, Weak Spells and Breaks

The season is not a continuous rain but an alternation of wet and dry spells.

StateMonsoon troughRainfallAssociated features
Activeat or south of its normal position, well markedwidespread, heavy over central India and the west coastlows and depressions from the Bay moving WNW; strong low-level westerlies along the west coast; offshore troughs and vortices
Weakless markedbelow normal but not suppressed everywhereweak westerlies along the west coast
Breakshifted north to the Himalayan foothills; may disappear from the surface chart over central Indialittle rain over central and north-west India; increased rain over the foothills, north-east India and parts of the south peninsula (Tamil Nadu)low-level westerlies over the peninsula weaken; offshore trough and vortex along the west coast absent

Rainfall-based definition. For objective work, IMD and research studies average daily rainfall over a core monsoon zone of central India and standardise it. A break is commonly defined as a spell of at least three consecutive days with the standardised anomaly below −1.0; an active spell as at least three consecutive days above +1.0. Breaks are most frequent in July and August and typically last a few days to a couple of weeks.

Why breaks happen: intraseasonal oscillations.

ModePeriodPropagationEffect
Monsoon intraseasonal oscillation (MISO) / boreal summer intraseasonal oscillation30–60 daysnorthward from the equatorial Indian Ocean to the foothills, roughly 1° of latitude per dayalternate active and break phases over India; linked to the Madden–Julian Oscillation
Quasi-biweekly mode10–20 dayswestward from the western Pacific and Bay of Bengalspawns and modulates lows and depressions

A break over India is often matched by enhanced convection over the equatorial Indian Ocean, and vice versa: the northward-moving band leaves India dry as the next band forms near the equator.


Withdrawal

IMD’s criteria

CriterionDetail
Earliest datewithdrawal from extreme north-west India is not declared before 1 September
Raincessation of rainfall over the area for 5 consecutive days
Circulationestablishment of an anticyclone in the lower troposphere (850 hPa and below)
Moistureconsiderable reduction in moisture, judged from water-vapour satellite imagery and tephigrams
Later stagesthe line of withdrawal then moves south and south-east; the normal date for withdrawal from the whole country is about 15 October

What happens physically

  • The sun moves south; the heat low weakens and the monsoon trough shifts south and fills.
  • An anticyclone builds over north-west India at low levels: dry north-westerlies, clear skies and a large diurnal temperature range.
  • The TEJ weakens and the STJ returns south of the Himalaya.
  • The “October heat” in north India: clear skies and still-high sun give hot, humid days after the rain stops.
  • The Bay of Bengal remains active: post-monsoon cyclones (October–December) are among the strongest of the year.

The Northeast Monsoon

FeatureDetail
SeasonOctober–December
Windsnorth-easterlies at the surface over the south peninsula and the Bay; they cross the Bay of Bengal and pick up moisture
RegionsTamil Nadu and Puducherry, coastal Andhra Pradesh, Rayalaseema, south interior Karnataka, Kerala
Importancecoastal Tamil Nadu gets nearly half its annual rainfall in this season; its rainfall peaks in October–November, not July
Onsetover coastal Tamil Nadu in the second half of October, after the southwest monsoon has withdrawn from the south peninsula
Rain-bearing systemseasterly waves, lows, depressions and cyclones from the Bay of Bengal; the ITCZ, now over the southern peninsula and the south Bay
ENSO linktends to be enhanced during El Niño and positive IOD years, the opposite sign to the southwest monsoon’s tendency

Northeast monsoon onset over Tamil Nadu is declared when the southwest monsoon has withdrawn from the south peninsula, persistent north-easterly or easterly winds have set in over the Tamil Nadu coast, and widespread rainfall has begun over coastal Tamil Nadu and adjoining areas. (IMD’s exact thresholds are not reproduced here.)

Why the east coast of Tamil Nadu is dry in July. In the southwest monsoon season it lies in the rain shadow of the Western Ghats, and the westerlies reaching it are descending. The NE monsoon reverses this geometry: moist easterlies from the Bay meet the coast directly.


Interannual Variability (in brief)

FactorTypical association with the southwest monsoon
El Niñotends to weaken it; many but not all El Niño years are deficient
La Niñatends to strengthen it
Positive Indian Ocean Dipoletends to favour Indian rainfall; can offset El Niño
Equatorial Indian Ocean oscillation (EQUINOO)enhanced convection over the western equatorial Indian Ocean tends to favour Indian rainfall
Eurasian and Himalayan snow cover in winter–springheavy snow tends to precede weaker monsoons (Blanford’s 19th-century idea)

Details and indices are in Monsoon, Cyclones and ENSO.


Memory Tricks

Don’t Confuse

ThisvsThat
Onset: abrupt, criteria-based, around 1 June over KeralavsPre-monsoon showers: local thunderstorms, not onset
Bogus onset: early heavy rain without the full circulationvsTrue onset: rain + deep westerlies + low OLR
Break: trough at the foothills, central India dryvsWeak monsoon: trough present but rainfall below normal
Hiatus: a pause in the advance (June)vsBreak: a pause after the monsoon has covered the country (July–August)
Withdrawal: end of the season from NW India, September onwardvsBreak: temporary, mid-season
SW monsoon: June–September, westerlies, whole countryvsNE monsoon: October–December, easterlies, south-east peninsula
El Niño: tends to weaken SW monsoonvsEl Niño: tends to enhance NE monsoon
MISO: 30–60 days, northwardvsQuasi-biweekly mode: 10–20 days, westward

Common Mistakes


Practice

All questions below are original practice questions written in CSIR NET style. They are not previous-year questions.

Level 1: Foundation

  1. What is the normal date of monsoon onset over Kerala, and from where does withdrawal begin?
  2. What OLR threshold does IMD use in its Kerala onset criteria?
  3. Which season gives coastal Tamil Nadu its main rainfall?
Answers
  1. 1 June; withdrawal begins from north-west India (normal date 17 September).
  2. Below 200 W m⁻² over 5–10° N, 70–75° E.
  3. The northeast monsoon, October–December.

Level 2: Part B Style

Q1 (practice). The all-India June–September rainfall in a year is 760 mm. With an LPA of 868.6 mm, the season is:

  • (a) excess
  • (b) normal
  • (c) deficient
  • (d) above normal
Answer and explanation

(c). 760/868.6 = 87.5% of LPA, a deficit of 12.5%. More than 10% below LPA is deficient. (For 800 mm, 92.1%, the season would be normal in the realised ±10% sense but “below normal” in IMD’s five-category forecast scheme.)

Q2 (practice). In mid-July the monsoon trough is seen along the Himalayan foothills and the surface chart shows no trough over central India. The expected rainfall pattern is:

  • (a) heavy rain over central India and the west coast
  • (b) little rain over central India; enhanced rain over the foothills and north-east India
  • (c) heavy rain over Rajasthan
  • (d) uniform moderate rain everywhere
Answer and explanation

(b). This is a break. Rain follows the trough to the foothills and the north-east; central India dries.

Q3 (practice). Which change in the upper troposphere accompanies the onset of the monsoon over north India?

  • (a) The subtropical westerly jet moves south of the Himalaya.
  • (b) The subtropical westerly jet moves north of the Tibetan Plateau and the tropical easterly jet sets in.
  • (c) The polar front jet strengthens over central India.
  • (d) The Somali jet rises to 200 hPa.
Answer and explanation

(b). The heating of Tibet reverses the upper meridional temperature gradient; the STJ moves north and the TEJ forms. Withdrawal reverses the switch.

Q4 (practice). A northward-propagating convective band moves about 1° of latitude per day. Roughly how long does it take to travel from 5° N to 25° N?

  • (a) 2 days
  • (b) 5 days
  • (c) 20 days
  • (d) 60 days
Answer and explanation

(c). 20° at 1° per day = 20 days, consistent with a 30–60-day oscillation in which successive bands form near the equator while the previous one reaches the foothills.

Level 3: Part C Style

Q1 (practice). The following data are available for late May of a year (14 listed onset stations).

DateStations with ≥ 2.5 mmWesterlies up to (hPa), box Eq–10° N, 55–80° E925 hPa zonal wind, 5–10° N, 70–80° E (kt)OLR, 5–10° N, 70–75° E (W m⁻²)
8 May1165018185
9 May1260019180
27 May970012215
28 May1065014205
29 May1060017195
30 May1155019180

On which date, if any, can onset over Kerala be declared? Justify each rejected date.

Answer and worked solution

Threshold: 60% of 14 = 8.4, so at least 9 stations on two consecutive days. Wind: westerlies to at least 600 hPa and about 15–20 kt at 925 hPa. OLR below 200 W m⁻².

  • 8–9 May: rain, wind and OLR all satisfied, but rainfall is counted only after 10 May. Not onset (a possible bogus onset).
  • 27–28 May: rain criterion met (9 and 10 stations), but westerlies reach only 700 and 650 hPa (not up to 600 hPa), 925 hPa winds are 12–14 kt, and OLR is above 200 W m⁻². Not onset.
  • 28–29 May: rain met on both days; on 29 May, westerlies reach 600 hPa, wind 17 kt and OLR 195 W m⁻² all satisfy the criteria. Onset is declared on the second of the two rainy days, so 29 May.

Onset: 29 May. (30 May also satisfies the criteria, but onset is declared on the first qualifying second day.)

Q2 (practice). The standardised daily rainfall anomaly over the core monsoon zone for 1–20 July is:

Day1234567891011121314151617181920
Anomaly1.21.41.10.6−0.3−1.1−1.3−0.9−1.2−1.5−1.4−1.1−0.40.21.01.31.60.8−1.2−1.3

Using the “at least three consecutive days” rule with thresholds of ±1.0, identify the active and break spells. What is the most likely position of the monsoon trough on day 11, and where would you expect above-normal rain then?

Answer and worked solution
  • Active spell: days 1–3 (1.2, 1.4, 1.1, three days above +1.0).
  • Days 6–7 are below −1.0 but day 8 (−0.9) interrupts: only two days, not a break.
  • Break: days 9–12 (−1.2, −1.5, −1.4, −1.1), four consecutive days below −1.0.
  • Days 15–17: 1.0, 1.3, 1.6. The rule needs values above +1.0; day 15 is exactly 1.0, so strictly only days 16–17 qualify (two days), not an active spell under a strict reading. (If the threshold were “≥ 1.0”, days 15–17 would qualify. Read the definition carefully in the exam.)
  • Days 19–20: only two days; the series ends, so undetermined.

On day 11 (mid-break) the trough is most likely along the Himalayan foothills; expect above-normal rain over the foothills, north-east India and parts of the south peninsula.

Q3 (practice). Which of the following statements are correct?

(A) Withdrawal of the southwest monsoon from extreme north-west India is not declared before 1 September and requires a lower-tropospheric anticyclone and about five dry days. (B) The northeast monsoon over Tamil Nadu tends to be enhanced in El Niño years. (C) A hiatus in June means the monsoon has withdrawn and will set in again. (D) An onset vortex over the east-central Arabian Sea can accompany the onset over Kerala.

  • (a) A, B and D only
  • (b) A and C only
  • (c) B and D only
  • (d) A, B, C and D
Answer and worked solution

(a) A, B and D.

  • (A) Correct statement of the withdrawal criteria.
  • (B) Correct; the sign of the ENSO link is opposite to that of the southwest monsoon.
  • (C) Incorrect: a hiatus is a stall in the advance; nothing is withdrawn.
  • (D) Correct.

Quick Revision Table

ItemFact
Normal datesKerala 1 June; whole country 8 July; withdrawal from NW India 17 Sept; complete about 15 Oct (revised 2020)
LPA868.6 mm (1971–2020); normal ±10%; forecast ’normal’ 96–104%
Onset criteriaafter 10 May; 60% of 14 stations ≥ 2.5 mm on 2 days (declare on 2nd); westerlies to 600 hPa; 15–20 kt at 925 hPa; OLR < 200 W m⁻²
Onset signalsKE jump in Arabian Sea westerlies; onset vortex; STJ north of Tibet; TEJ sets in
AdvanceArabian Sea and Bay branches; mid-June hiatus common
Activetrough normal or south; depressions; strong west-coast westerlies; offshore vortices
Breaktrough at foothills; dry central India; rain over foothills, NE India, south
Break definitionstandardised anomaly below −1.0 for ≥ 3 days over the core monsoon zone
MISO30–60 days, northward about 1° per day; quasi-biweekly 10–20 days, westward
Withdrawalnot before 1 Sept; 5 dry days; 850 hPa anticyclone; moisture drop
NE monsoonOct–Dec; NE winds; Tamil Nadu nearly half its annual rain; onset: SW monsoon withdrawn from south peninsula, persistent easterlies, widespread rain
ENSOEl Niño weakens SW monsoon, tends to enhance NE monsoon

Key Takeaways

What’s Next